The long-term objective of this proposal is to understand how the brain learns and control movement sequences. As Lashley recognized more than half a century ago, much of our behavior, from the performance of organized movements to the ability to use language, is based on our capacity to detect, learn, and produce sequences. In the current proposal, we use variants of the serial reaction time (SRT) task and functional imaging in human subjects to examine the neural substrates responsible for learning the fundamental structure of movement sequences, the brain areas responsible for modulating learning through reward and punishment, and the extent to which the brain uses similar strategies for learning temporal and spatial sequences. We will test the following hypotheses. (1) During sequence learning cortical motor areas detect and learn transitions from one element to the next, while the basal ganglia encode the whole structure of sequenced movements. (2) Reward and punishment have direct but differential effects on motor sequence learning and this will be reflected by proportional changes in the activity of the basal ganglia. (3) Learning sequences of temporal intervals will engage a similar set of brain areas to those involved in learning spatial sequences and will not involve the cerebellum. Impairment in the ability to produce sequences is an important component of the disability experienced by patients with Parkinson's disease. The work outlined here will provide a fundamental understanding of these disabilities and may lead to the development of strategies for rehabilitation and treatment of these patients.